Millions of people participate in online science projects. Foldit takes things to a new level, using an online game to leverage the forces of human creativity, intuition and spatial reasoning against a scientific task that flummoxes silicon brains.
Susanne Halicki’s last science class was in high school — and that was a long time ago.
She’s got grown children now and works as an office administrator.
But every evening, Halicki slips on a cyber lab coat and immerses herself in molecular biology via a video game developed in Seattle.
The British woman is among 150,000 people worldwide who have tried their hand at the game called Foldit. She’s also part of an elite subset of players who have become so skillful at sussing out the structures of biologically important proteins that they frequently outperform supercomputers.
- Nathan Hale High School juniors boycott state test
- Scientists to study the 'modern miracle' of Ozzy Osbourne's survival
- Jesse Jones is back: Seattle's superhero consumer reporter is now at KIRO 7
- Seahawks' toughness is not for everyone
- Ditching Dreamliners: United buys older, cheaper planes
Most Read Stories
“I like solving puzzles,” Halicki said. “It was hard at first, but I persevered.”
Insights from the citizen scientists are already helping in the quest for new drugs and green-energy technologies, and could become an even more powerful tool in the future, said University of Washington biochemist David Baker.
“This shows that people who don’t have a scientific background can solve very challenging scientific problems,” said Baker, co-creator of Foldit. He’s now enlisting the players to design potential drugs from scratch.
Foldit’s initial success, reported last week in the journal Nature, also gives a boost to fledgling efforts to harness the collective brain power of amateurs. Many skeptics had dismissed “distributed thinking” projects as little more than PR stunts.
“I think this is the start of something really big,” said Zoran Popovic, a UW computer scientist who collaborated with Baker and several colleagues on Foldit. “Think of how many other problems we could possibly solve with this … completely new way to do science.”
Millions of people participate in online science projects. The simplest are screen savers that devote idle computer time to automated tasks, such as scanning radio telescope data for alien transmissions. More demanding projects rely on volunteers to sort through millions of images, categorizing galaxies or searching for particles from beyond the solar system.
Foldit takes things to the next level, using an online game to leverage the forces of human creativity, intuition and spatial reasoning against a scientific task that flummoxes silicon brains.
The game’s name derives from the way proteins naturally fold into three-dimensional shapes inside living cells. The molecular machines behind the vital functions of life, proteins control everything from digestion to healing and immunity. But the molecules don’t function without the proper shape.
The problem for scientists trying to understand how the body works and cure disease is that any single protein chain can theoretically twist and fold into millions of possible shapes.
Baker and his team developed a computer program called Rosetta to sort through that mind-boggling array and predict a protein’s structure based on its building blocks. A screen-saver version, Rosetta@home, taps into personal computers for the heavy number crunching.
But lacking human intuition and visualization skills, Rosetta makes a lot of dumb mistakes. So Baker and Zoran conceived Foldit to see if people could educate the model. Players compete for top scores and rankings, using their cursors to push, pull and wiggle color-coded protein chains. Stable configurations earn points. Arrangements that violate the laws of physics trigger warnings and subtract from the score.
“I try to see it as a whole picture, to see what I can change to make it fit,” said player Mark Thompson. Hooked while recovering from heart surgery, the 47-year-old Brit credits his experience as a construction worker for sharpening his ability to visualize in three dimensions.
Even without the trolls and weapon-wielding warriors of other games, Foldit hooks its loyal players.
“I’m not addicted,” Thompson said with a laugh. “I just can’t stop playing.”
Working first with molecules of known structure, the top protein wranglers — most of whom organized into teams — bested Rosetta in many cases. Their track record held up when they tackled a set of proteins whose structures were unknown.
Few of those who top the leaderboard have any scientific training. But most share Thompson’s spatial skills and Halicki’s persistence, Popovic said. He and his colleagues are dissecting the players’ most-effective folding strategies to improve Rosetta.
Other citizen-science efforts are also yielding modest results.
Volunteers for Stardust@home discovered three grains of what may be interstellar dust by poring over images from the Stardust spacecraft. Astronomers are abuzz about a strange green blob spotted by a volunteer scanning images for a project called GalaxyZoo.
Popovic believes the next generation of what he calls scientific discovery games will lead to a “democratization” of science, where the contributions of nonexperts will inform and advance the work of professionals.
David Anderson, director of SETI@home — the alien-searching screen saver that kicked off the craze — is working on a new program to allow more scientists to recruit amateurs to help with their work.
“The human brain can do a lot of things better than computers,” he said. “The question of how best to harness that for scientific purposes hasn’t been worked out yet,” he said.
Sandi Doughton: 206-464-2491 or email@example.com